Hysteresis Compensation in Force/Torque Sensor based on Machine Learning

This paper proposes a method to improve the accuracy of the force/torque (F/T) sensor based on machine learning considering time series data. There are several problems with F/T sensors, one of which is hysteresis. Hysteresis is a factor of error dependent on force history. There have been few researches focusing on hysteresis in an F/T sensor. We solved this problem by considering time series data. Time series data was put into machine learning such as linear regression and Support Vector Regression (SVR). We evaluated this method with an existing high dynamic range F/T sensor. We confirmed that the error decreased in both high and low force ranges. Since there is nonlinearity in hysteresis, we predicted that SVR will be more accurate than linear regression. Linear regression considering time series was better than SVR when loading training data at random intervals and loading test data at constant intervals.

[1]  James Griffin,et al.  Traceability of Acoustic Emission measurements for a proposed calibration method – Classification of characteristics and identification using signal analysis , 2015 .

[2]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[3]  Shakeb A Khan,et al.  Sensor calibration and compensation using artificial neural network. , 2003, ISA transactions.

[4]  Kouhei Ohnishi,et al.  Frequency-based analysis of the relationship between cutting force and CT number for an implant-surgery-teaching robot , 2017 .

[5]  Xinong Zhang,et al.  Hybrid calibration method for six-component force/torque transducers of wind tunnel balance based on support vector machines , 2013 .

[6]  Hyoukryeol Choi,et al.  Force/torque sensor calibration method by using deep-learning , 2017, 2017 14th International Conference on Ubiquitous Robots and Ambient Intelligence (URAI).

[7]  Ryogo Kubo,et al.  Human-Machine Cooperative Grasping/Manipulating System Using Force-based Compliance Controller with Force Threshold , 2016 .

[8]  Tien-Fu Lu,et al.  Neural-network-based 3D force/torque sensor calibration for robot applications , 1997 .

[9]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[10]  Minoru Asada,et al.  Contact force estimation from flexible tactile sensor values considering hysteresis by Gaussian process , 2014, 4th International Conference on Development and Learning and on Epigenetic Robotics.

[11]  Shigeki Sugano,et al.  Design and Characterization of a Three-Axis Hall Effect-Based Soft Skin Sensor , 2016, Sensors.

[12]  Hong Liu,et al.  DLR-Hand II: next generation of a dextrous robot hand , 2001, Proceedings 2001 ICRA. IEEE International Conference on Robotics and Automation (Cat. No.01CH37164).

[13]  Toshiaki Tsuji,et al.  Development of a multistage six-axis force sensor with a high dynamic range , 2017, 2017 IEEE 26th International Symposium on Industrial Electronics (ISIE).

[14]  Aiguo Song,et al.  Design and Calibration of a Six-axis Force/torque Sensor with Large Measurement Range Used for the Space Manipulator , 2015 .

[15]  Yongjun Sun,et al.  Analysis calibration system error of six-dimension force/torque sensor for space robot , 2014, 2014 IEEE International Conference on Mechatronics and Automation.